These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Out of Pod Experience

 
  • Topic is locked indefinitely.
 

(As Promised) Why Statistics Are Pseudoscience
Author
Previous Topic Next Topic
Eternum Praetorian
Doomheim
#141 - 2012-08-07 22:16:24 UTC  |  Edited by: Eternum Praetorian
Akita T wrote:
Let's test the ABSURDITY of your approach with a PRACTICAL example.

Tell me, in your opinion, what is the chance to get a streak of 2 heads when you make 200 throws ?
According to your previously stated approach you took for the 7head-20throw case, it should be 198 : 2^200 = almost zero.

Now throw a coin 200 times and tell me you didn't manage to get at least 2 heads in a row at least once.
What do you know, something your math says should almost never happen manages to happen almost always.
Obviously your approach is flawed.



2^200 is 1,606,938,044,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000

So then you plug in (2 -1,606,938,044,000,000,000,000,000,000,000,000,000,000,000,000,000,
000,000,000,000) + 1 = 1,606,938,044,999,999,999,999,999,999,999,999,999,999,999,999,999,
999,999,999,999.

So the odds of getting heads at least twice in 2 hundred flips is a truly absurd

1.6069380449x10 -59 : 1.6069380440 -59 ... which sounds about right to me.

[center]The EVE Gateway Blog[/center] [center]One Of EVE Online's Ultimate Resources[/center]
Eternum Praetorian
Doomheim
#142 - 2012-08-07 22:20:02 UTC  |  Edited by: Eternum Praetorian
FloppieTheBanjoClown wrote:
Eternum Praetorian wrote:
Those are simple and short patterns mr Floppie. This is about long repetitious events being consistently generated from random events. Like: 1111111111111111111^2 or 00000000000000000000^2 or 101010101010101010^2 or 11001100110011001100^2 or 111000111000111000111000^2


What's amusing about this is that you think the sequence of a specific thousand bits is significant among a billion billion billion of them. At a scale like that, a thousand 1s in a row is insubstantial. Also...you do understand the concept of examples, right?



I do not think so. I am not talking about a billion billion billion. I am talking about a few hundred to a few thousand and in a few hundred, samples like that they are indeed very unique.

[center]The EVE Gateway Blog[/center] [center]One Of EVE Online's Ultimate Resources[/center]
dexington
Caldari Provisions
Caldari State
#143 - 2012-08-08 04:49:54 UTC
Eternum Praetorian wrote:

So then you plug in (2 -1,606,938,044,000,000,000,000,000,000,000,000,000,000,000,000,000,
000,000,000,000) + 1 = 1,606,938,044,999,999,999,999,999,999,999,999,999,999,999,999,999,
999,999,999,999.


it is a negativ number:
(2 -1,606,938,044,000,000,000,000,000,000,000,000,000,000,000,000,000,
000,000,000,000) + 1 = -1,606,938,043,999,999,999,999,999,999,999,999,999,999,999,999,999,
999,999,999,999

I'm a relatively respectable citizen. Multiple felon perhaps, but certainly not dangerous.
Akita T
Caldari Navy Volunteer Task Force
#144 - 2012-08-08 06:05:31 UTC  |  Edited by: Akita T
Eternum Praetorian wrote:
So the odds of getting heads at least twice in 2 hundred flips is a truly absurd

You were using a different type of calculation, and you even got that wrong calculation wrong in a different way.
Also, you actually listed about 10:1 odds, not "truly absurd" odds.
And yet you almost always get them (as in, 99.99+% of the time), not just 90% of the time.

Either way, you just admitted you're completely and utterly wrong from every possible viewpoint.

Let's draw this to a conclusion.

Statistics are only a pseudoscience to you if you have next to absolutely no clue whatsoever how to properly calculate stuff.
So, yeah, since you have repeatedly proven you can't even understand a detailed explanation of how the proper calculations are made, let alone do them yourself from scratch, obviously, it's no wonder that TO YOU, statistics looks like pseudoscience.
For the rest of the people, those that can at least follow the explanations if nothing else, statistics is quite scientific, thankyouverymuch.

Akita T evewiki profile | | | Build your own EVE-ready PC
Eternum Praetorian
Doomheim
#145 - 2012-08-08 11:50:12 UTC  |  Edited by: Eternum Praetorian
I did screw up on that last one I admit. But the most compelling evidence of my theory came from you. You just don't like that Idea very much Big smile


I was supposed to do (2-200) + 1 = 199 combinations of two in a row (face palm). This value is referring to two in a row and only two in a row. Not 3 in a row or 4 in a row. Two in a row in 200.



So it would be 199:200 which is (in a round about) way what I said the first time. You need to stop changing the subject and refer back to your previous proofs. I will explain it one last time.



Quote:

1. If there has not been enough time in the universe to flip 100 heads in a row

THEN

2. It is unreasonable to believe that you will see that event during your short stay in a casino.




This phenomena is explained when you understand how the perspective of the observer works. I get this idea is flying in the face of your "math religion" but it does make sense, the math does support it and no amount of ad hominem is going to change that. The best math supporting it actually came from you, so I thank you for the proofs. They work quite well. Cool



Don't be angry!

[center]The EVE Gateway Blog[/center] [center]One Of EVE Online's Ultimate Resources[/center]
Eternum Praetorian
Doomheim
#146 - 2012-08-08 11:57:04 UTC
Hmmm... perhaps it is time for a new thread. I was wrong statistics are not pseudoscience they are just "not done". Perhaps a new thread explaining relativity is in order.


Mwhahahaah!!!

[center]The EVE Gateway Blog[/center] [center]One Of EVE Online's Ultimate Resources[/center]
Akita T
Caldari Navy Volunteer Task Force
#147 - 2012-08-08 12:24:08 UTC  |  Edited by: Akita T
You're still calculating the 2 flips thing all wrong, but, eh, nevermind.

About the 100 flip streak thing, gee, so you finally came to the exact conclusion we've been telling you is the correct one and you kept on resisting until now ?
Of course you're personally unlikely to see a streak of 100 (or even just 30) in a short trip to a casino, that's the freaking point of "it's only likely if you get an insane number of trials total".
Nevertheless, a streak of 30 or 100 or any other length is eventually going to become very likely if you just could get enough time to make enough trows (you can't in a single human lifetime, not for noticeably more than 30-ish throws, even if that is all you ever do your entire life), it's only about the number of throws you are allowed to make and nothing else, a number of throws which you kept insisting we should consider as good as infinite if we wanted to.
So you belatedly came to a conclusion that is the same as what we told you near the beginning of the first thread you started, and now you want to start a third thread... about what exactly ?

P.S. And you're also using the term "relativity" in a really inappropriate way.
You want to use "comparatively" or something else with a similar meaning in the cases in which you were usually employing it so far.
You're confusing people because they are first thinking you're talking about Einstein's theory of relativity, which is something completely and utterly different from what you're talking about.

Akita T evewiki profile | | | Build your own EVE-ready PC
stoicfaux
#148 - 2012-08-08 13:22:43 UTC
Eternum Praetorian wrote:
Hmmm... perhaps it is time for a new thread. I was wrong statistics are not pseudoscience they are just "not done". Perhaps a new thread explaining relativity is in order.


Mwhahahaah!!!

Eh... I wouldn't recommend it. The problem with using math/science when trolling is there is often a correct answer, well defined scenarios, rigid proofs, etc., so instead of harvesting tears and rage, you normally wind up looking stupid (because you're provably wrong) and otherwise discouraging people from taking up teaching as a profession.

Not that I would know this from personal experience of course.

Pon Farr Memorial: once every 7 years, all the carebears in high-sec must PvP or they will be temp-banned.
Akita T
Caldari Navy Volunteer Task Force
#149 - 2012-08-08 13:28:38 UTC
Maybe he should be talking about genetics instead next. It's loads of statistics too, but people still usually get it wrong because of the other thing involved :p

Akita T evewiki profile | | | Build your own EVE-ready PC
Eternum Praetorian
Doomheim
#150 - 2012-08-08 14:26:20 UTC
Akita T wrote:
Maybe he should be talking about genetics instead next. It's loads of statistics too, but people still usually get it wrong because of the other thing involved :p



Ah yes because the mystical quality of randomness resulted in all life as we know it. Winning!

[center]The EVE Gateway Blog[/center] [center]One Of EVE Online's Ultimate Resources[/center]
Akita T
Caldari Navy Volunteer Task Force
#151 - 2012-08-08 14:53:30 UTC  |  Edited by: Akita T
Eternum Praetorian wrote:
Ah yes because the mystical quality of randomness resulted in all life as we know it. Winning!

More like the quick cut-out of unuseful randomness, but yeah, basically, sort of randomness.
And there's your new thread - go, go, GO !!! Twisted

Akita T evewiki profile | | | Build your own EVE-ready PC
FloppieTheBanjoClown
Arcana Imperii Ltd.
#152 - 2012-08-08 15:01:31 UTC
Eternum Praetorian wrote:
Hmmm... perhaps it is time for a new thread. I was wrong statistics are not pseudoscience they are just "not done". Perhaps a new thread explaining relativity is in order.

Mwhahahaah!!!


While you're at it, impress us with phrases like "just a theory".

I'm stumped, though. I can't tell whether you're a very persistent and talented troll, or you actually believe the tripe you're pushing here. I only stick around because it's a decent exercise in argument.

Founding member of the Belligerent Undesirables movement.
Akita T
Caldari Navy Volunteer Task Force
#153 - 2012-08-08 16:16:09 UTC
He's almost certainly a troll, not a very good one, but not bad either, and somewhat entertaining.

Akita T evewiki profile | | | Build your own EVE-ready PC
Eternum Praetorian
Doomheim
#154 - 2012-08-08 23:47:13 UTC
Akita T wrote:
He's almost certainly a troll, not a very good one, but not bad either, and somewhat entertaining.



Proofs of concept as requested. Cool
https://forums.eveonline.com/default.aspx?g=posts&t=141944&find=unread

[center]The EVE Gateway Blog[/center] [center]One Of EVE Online's Ultimate Resources[/center]
Stigman Zuwadza
Deep Core Mining Inc.
Caldari State
#155 - 2012-08-09 03:32:33 UTC  |  Edited by: Stigman Zuwadza
I promise I tried to read through this thread but my attention span for reading is somewhat short. It occured to me that a fundemental thing was maybe missed. That being that...

The 50/50, 1:2 and so on refers to the likelihood of it being 1 of the 2 sides and not to the probability of it being a particular side.

This being the case the 50/50 idea holds up as being true (to me) in that with each flip of the coin 1 of the sides was the resultant face, so this means that every flip does in fact have a 50/50 chance of it being 1 of the 2 faces. It may be that this idea is somehow metamorphosed into becoming the probability of it being a certain face when in fact it refers merely to the fact that it is 1 of the 2 sides.

To me the whole probability of it be a particular face is another matter altogether.

My 0.00000000000000001 isks worth. Big smile

Fly safe. o7

It's broken and it's been broken for a long time and it'll be broken for some time to come.
Kenneth O'Hara
Sebiestor Tribe
#156 - 2012-11-06 06:18:54 UTC
=recent&filters[primary]=images]Thread Necromancy!!! Hazzah!!!

Bring Saede Riordan back!! Never Forget! _"__Operation Godzilla Smacks Zeus"  ~__Graygor _